Improvement of Conflict Detection and Resolution at High Densities Through Reinforcement Learning

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

54 Downloads (Pure)

Abstract

The use of drones for applications such as package delivery, in an urban setting, would result in traffic densities that are orders of magnitude higher than any observed in manned aviation. Current geometric resolution models have proven to be very efficient. However, at the extreme densities envisioned for such drone applications, performance is hindered by unpredictable emergent behaviour of interacting traffic. This paper describes a study that intends to investigate how reinforcement learning techniques can be used to complement geometric methods, thus improving conflict detection and resolution at high traffic densities. Different hybrid approaches are discussed, and preliminary results are shown for a hybrid model that uses geometric methods in the training phase of a Deep Deterministic Policy Gradient (DDPG) model.
Original languageEnglish
Title of host publicationICRAT 2020
Number of pages4
Publication statusPublished - 2020
EventICRAT 2020: International conference on research in air transportation - Virtual/online event due to COVID-19
Duration: 15 Sept 202015 Sept 2020
Conference number: 9
http://www.icrat.org/icrat/upcoming-conference/committee-members/

Conference

ConferenceICRAT 2020: International conference on research in air transportation
Abbreviated titleICRAT 2020
Period15/09/2015/09/20
Internet address

Bibliographical note

Virtual/online event due to COVID-19

Keywords

  • Conflict Detection and Resolution (CD&R)
  • reinforcement learning (RL)
  • Deep Deterministic Policy Gradient (DDPG)
  • Modified Voltage Potential (MVP)
  • U-Space
  • Unmanned Traffic Management (UTM)
  • Self-Separation
  • BlueSky ATM Simulator

Fingerprint

Dive into the research topics of 'Improvement of Conflict Detection and Resolution at High Densities Through Reinforcement Learning'. Together they form a unique fingerprint.

Cite this